Glioblastoma (GBM) is a devastating brain cancer with a mean survival of shorter than 2 years. Current standard- of-care therapies provide only palliation, indicating an urgent need to develop more effective therapies. Accumulating evidence suggests that GBM stem-like cells (GSCs) possess intrinsic molecular mechanisms and adopt extrinsic signals to escape from therapeutic insults, causing subsequent lethal tumor recurrence. However, molecular and cellular mechanisms that regulate how GSCs react to therapeutic insult remain poorly understood. Thus, this knowledge gap has hampered efforts to develop effective therapies that prevent GBM growth and recurrence. In this proposal, we will test the hypothesis that therapy-induced adoptive changes in immune microenvironment promotes GSC therapy resistance to drive GBM recurrence. We have developed this hypothesis based on our recent findings in our 6 research articles. First, we identified that the mesenchymal differentiation of GSCs promotes therapy resistance accompanied by the aberrant NF-kB activation and increased pro-tumorigenic tumor-associated macrophages (TAMs) (Mao et al., PNAS 2013; Bhat et al., Cancer Cell 2013; Kim et al., Cancer Cell 2016). Following these studies, we identified that the mesenchymal GSCs activates the receptor tyrosine kinase AXL (Chang et al., Stem Cell Reports 2015), unlike the non-Mesenchymal GSCs with activated PDGFR? (Jeon et al., Cancer Res 2014; Sadahiro et al., Cancer Res 2018). To further extend these studies, we will investigate if and how the crosstalk between mesenchymal GSCs and TAMs promotes therapy resistance of GBM. In Aim 1, we will examine the molecular mechanisms in GSCs that receive inter-cellular signals from TAMs through AXL. In Aim 2, we will examine the molecular signaling mechanisms in TAMs that increase secretion of the AXL-binding ligand PROS1 after immunotherapy with Nivolumab (PD-1 inhibitor). In Aim 3, we will determine the efficacy of the combination of Nivolumab and the AXL inhibitor BGB324 for the preclinical GBM models. Collectively, we anticipate that this study will yield a new paradigm for the GSC biology and a novel therapeutic approach to target key regulators of GSC, which may lead to the translation into improved therapies.